AUTOTHERMAL REACTOR CONCEPT FOR COMBINED OXIDATIVE COUPLING AND METHANE REFORMING
T.P. Tiemersma, M. van Sint Annaland, J.A.M. Kuipers
Faculty of Science and Technology - Fundamentals of Chemical Reaction Engineering, University of Twente, P.O. Box 217, 7500AE Enschede, The Netherlands
A novel autothermal reactor concept has been developed for the simultaneous production of ethylene by oxidative coupling (OCM) and synthesis gas by steam reforming of methane (SRM), supported by a detailed numerical modeling study. The proposed reactor consists of two separate reaction chambers which are thermally coupled. The OCM is carried out in packed bed reverse flow membrane reactor tubes submerged in a fluidized bed where the unconverted methane and by-products, from which the valuable C2 components have been
separated, are reformed together with some additional steam producing synthesis gas and consuming the reaction enthalpy emerged from the exothermic OCM.
Keywords: Chemical Reactors, Reverse Flow, Membranes, Autothermal Operation, Dynamic Simulation, Reaction engineering
1. INTRODUCTION
Bulk chemicals like ethylene with an annual production of 20,000-30,000 metric kilotons in Europe are currently mainly produced from crude oil via a cracking route. With the rapid consumption of the world oil reserves, it is expected that natural gas will more and more become a widely used source of chemicals. The indirect conversion route involves the energy-intensive production of synthesis gas and the consecutive production of higher hydrocarbons via the Fischer-Tropsch process and downstream endothermic steam cracking, resulting in a process with a large methane recycle, a lot of unit operations and reduced energy efficiency. More interesting is the direct conversion of natural gas into ethylene and ethane, which can be achieved by oxidative coupling of methane (OCM), at relatively high temperatures (>750 °C). Despite large efforts from many researchers, the maximum single-pass C2
yield on the best catalysts is limited to only about 15-20% due to undesired combustion reactions, deteriorating the selectivity to C2’s. However, if the mixture of by-products from the OCM reactions, typically CO, CO2, H2 and H2O
(see Fig. 1), could be shifted and reformed with methane to synthesis gas (SRM), two major advantages can be achieved.
First, the overall methane conversion to valuable products can be increased, producing C2’s (reaction 1) and
synthesis gas (reaction 3) simultaneously, where the produced synthesis gas can be used for the production of synthetic fuels via the Fischer-Tropsch process. Second, by combining the oxidative coupling and steam and dry reforming of methane an overall autothermal process can be achieved. The challenge is to introduce a reforming catalytic activity in such a way that the reaction rates of the exothermic OCM and the endothermic reforming reactions are balanced to create an autothermal and near isothermal process. This can be either performed on the scale of a single catalyst particle, resulting in the most efficient energy integration possible, or on reactor scale. In this paper it is investigated by a detailed modeling and experimental study whether the OCM and reforming reaction rates can be tuned, resulting in a novel reactor concept for the simultaneous and autothermal production of synthesis gas and higher hydrocarbons.
2. REACTOR CONCEPT
This paper concerns the thermal balancing of the heat released from the exothermic OCM reaction with the endothermic SRM between two different reactor compartments. Both reactions can be performed in several different reactor types, the optimal reactor concept highly depends on the type of process (i.e. exothermicity, reaction kinetics and thermodynamics). In a micro-catalytic fixed bed reactor, the reaction rates of the primary OCM reactions (reaction 1, 2) were determined on a Mn/Na2WO4/SiO2 catalyst at 800°C. The overall reaction order n towards O2
was determined to be lower for the OCM reaction (n=0.4) than for the methane combustion reaction (n=1). The lower reaction order towards O2 results in an increase of the overall C2 selectivity (as high as 85% at pO2 < 5 kPa) at
the expense of a lower CH4 conversion when the O2 partial pressure is decreased. In a conventional co-feed fixed
bed reactor it is not possible to obtain high C2 selectivity and a high CH4 conversion at the same time because of the
applied low O2 concentration.
By using distributive feed of O2/air the C2 selectivity can be enhanced because of the low O2 concentration while
feeding sufficient O2 along the reactor length to achieve a higher CH4 conversion than co-feed reactors. The
advantageous effect of distributed feed of O2/air has been shown with experiments in an isothermal packed bed
membrane reactor, where porous Al2O3 membranes were applied to distributively feed air to a diluted CH4 stream
over a Mn/Na2WO4/SiO2 catalyst at 800 °C. The total C2 selectivity increased from 20% to 30% at a CH4 conversion
of approximately 60% (see Fig. 2).
Distributed Pre-mixed 0 20 40 60 80 [%] CH4/O2/N2 = 1/1/14
CH4 conversion C2 selectivity C2 yield
Fig. 2. Effect of mode of distribution of O2 on performance of OCM in a fixed bed membrane reactor
(T=800°C, p=2 bar, mcat = 2.4 g).
Additionally the reaction heat is released along the entire reactor length, leading to a more uniform temperature profile. Still, unselective gas phase reactions can occur downstream at the reactor outlet, hence the product stream needs to be quenched to low temperature. For this reason the flow direction is periodically reversed so that both product and feed can be at low temperature and recuperative heat exchange between feed and product stream is integrated inside the fixed bed membrane reactor.
Fig. 3. Combination of a reverse flow packed bed membrane reactor (OCM) and fluidized bed reactor (SRM). In conventional autothermal reverse flow systems where the endo- and exothermic processes are carried out in counter- or cocurrently operated fixed bed heat exchange reactors, stable reactor operation (i.e. cyclic steady state) is difficult to achieve and to maintain because the temperature and concentration fronts of both reactions are traveling at different velocities, requiring complex process control. Next to difficult operability the temperature in the OCM system will increase to very high temperatures at which C2 selectivity decreases to undesired values. Hence efficient
cooling is required, which is achieved by immersing the OCM membrane reactor tubes in a fluidized bed reactor in which the reforming of the by-products of the OCM process is performed (see Fig. 3). The excellent mixing properties of the fluidized bed operation ensure a uniform operating temperature, so that the reaction heat produced by the exothermic reactions is efficiently removed. Additionally expensive high temperature (800 °C) gas heat exchangers are not necessary in this process, which can be advantageous from an economical point of view. The feasibility of the reverse flow OCM process is investigated by means of a 1D membrane reactor model, which solves the multi-component mass and energy balances taking into account convection, dispersion and chemical reactions. The influence of distributive feed of O2/air, operating temperature of the fluidized bed and heat transfer coefficients
on the overall reactor performance have been investigated. The performance of the SRM fluidized bed reactor, with the OCM product outlet composition (without C2 products) as the inlet composition, is evaluated and the required
reaction heat is correlated to the heat produced in the OCM fixed bed membrane reactor to determine effect of parameters on autothermal operation.
With the simulations it has been shown that indeed the exothermic OCM and endothermic SRM can be very efficiently coupled into an overall autothermal reactor and that in the cyclic steady state C2 yields up to 30% at full
CH4 conversion can be achieved with a CH4/O2 ratio of 2–2.5 and a H2O/CH4 ratio of 3 in the SRM fluidized bed
reactor.
